Generating non-Gaussian rough surfaces using analytical functions and spectral representation method with an iterative algorithm

Abstract

The non-Gaussian rough surface simulation method with desired spatial distribution and height distribution is generally used to analyse the contact characteristics of rough surfaces under different contact conditions. Conventional surface simulation methods have disadvantages in terms of their range, accuracy, and stability. In this study, the analytical function method is enhanced to generate non-Gaussian random number matrices. The enhanced method was combined with the spectral representation method and an iterative algorithm to accurately and stably generate rough surfaces characterized by extensive skewness, kurtosis and autocorrelation lengths. The skewness and kurtosis range of the generated rough surface includes skewness and kurtosis of most engineering surfaces, such as worn surfaces and various machined surface and irregular engineering surfaces. A rough surface is easily generated ≤ 10 s.